Movie Genre Classification Based on Arabic Synopsis
No Thumbnail Available
Date
2021
Authors
Mansour, Ahmad
Barham, Mahdi
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Text classification is a popular subject in machine learning. It’s
the process of assigning predefined labels to some text. However,
most of the research is directed towards the English language. Other
languages like Arabic feel neglected as the research on them is not
as extensive. When it comes to movie genre classification, which is
a subset of this field, it has been implemented in multiple ways for
the English language. However, in Arabic, the research is almost non existent. Normally, movie genre classification would require a human
to manually assign genres based on the synopsis, which can be time
consuming and inaccurate due to bias and human error. It is very
common for people to forget the name of a movie but remember the
general synopsis. It is also very common for people to want a movie
based on certain events. Thus, in this project, we are going to use
genre classification as a recommendation tool. As we will recommend
the top movies for the resulting genres. We are going to build a
website, that takes in the Arabic synopsis for a movie as input and
then use different machine learning algorithms to predict the genre of
the movie. Then we will recommend movies based on the genres. The
website will also have a login system and a profile to store each user’s
prediction history.